Talk:Deep Blue versus Kasparov, 1996, Game 1

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

THE DAY THAT I SENSED A NEW KIND OF INTELLIGENCE[edit]

Why is there no mention of this article by Kasparov on game 1 on this move? http://www.time.com/time/magazine/article/0,9171,984305,00.html 173.252.35.134 (talk) 12:48, 10 February 2012 (UTC)[reply]

When was this game played ?[edit]

The game was played on February 2, 1996 in Philadelphia, Pennsylvania ? On Deep Blue, the date was February 10, 1996. Can someone confirm this, please ? -- PFHLai 09:25, 2005 Feb 7 (UTC)

According to Chessbase's Megabase 2004, it was the 10th. I'll edit accordingly. --Camembert
Thanks. I also found the same February 10th date on http://www.uschess.org/results/tnmt/96kdb/game1.html and http://www.queensac.com/deepblueday.html . The event will be featured on the MainPage as a selected anniversary on this date. Thanks, again. -- PFHLai 17:44, 2005 Feb 7 (UTC)

On this link: http://en.wikipedia.org/wiki/Deep_Blue_-_Kasparov,_1997,_Game_6, it says that Kasparov won the first game: "Kasparov had won the first game, lost the second game (after resigning in a drawn position) and drawn games 3, 4 and 5 after having advantageous positions in all three. He was tired and dejected before this game." This page says Kasparov lost the first game. http://www.research.ibm.com/deepblue/

IN 1997 NOT 1996 !!!!!!! Lorenzo Alali french wikipedia user

Slant about rematch[edit]

The second paragraph makes several unsupported POV statements, including

  • The increase in strength between Deep Blue and "Deeper" Blue was minimal (the main article on Deep Blue says that several upgrades were made)
    • It doubled its power from 100,000,000 pos/s to 200,000,000 pos/s calculations, among other strength adjustments and was noted that Deep Blue 1997 would beat Deep Blue 1996. Information on this may be found in "Behind Deep Blue: Building the Computer that Defeated the World Chess Champion" book. Xonatron (talk) 01:40, 17 May 2017 (UTC)[reply]
  • Kasparov suffered a bout of paranoia (I remember this from reports at the time, but citations are needed)
  • That IBM retired Deep Blue because it had scored its publicity coup (Quite likely IBM made a calculated decision that the risk of losing wasn't worth any benefits of winning again, but this position would need to be supported).

I don't have a concrete suggestion for fixing this (or I'd have done so :-), but the main Deep Blue article takes a more even-handed view. Given this, much of the material can probably be removed. -Dmh 16:22, 5 December 2006 (UTC)[reply]

NOT a copyvio[edit]

Note that [1], linked to from the bottom of this page, was written by the same person as the original author of this article. There is no copyright infringement. --Camembert

It's been noted that much of this text is based on http://www.dwheeler.com/misc/deepblue-kasparov.txt - but this is not a copyright violation. I am the author of the text on that page, and I have contributed that text to Wikipedia. I record that information here (under "discussion") so that it won't be accidentally deleted speedily as a copyvio. I've modified the web page location noting its contribution to Wikipedia, to make sure that's clear. Enjoy! -- Dwheeler 18:10, 6 December 2006 (UTC)[reply]

Thanks for clearing it up Dwheeler. Sorry for the speedy tag. After 3 years most of your analysis still stands! Sander123 20:49, 6 December 2006 (UTC)[reply]
Pity the anlysis was so poor. Where did question mark on 25...Kh8? come from. It should, if anything be an ! and then followed by 27...f4! (27...d4 was inaccurate) after which black is better, with a draw likely. SunCreator (talk) 00:05, 25 October 2009 (UTC)[reply]

Confusing and OR[edit]

I realize something like this is bound to get very technical, but could it perhaps get simplifed some? I found myself completely lost in trying to read through the jargon. Also, statements such as "This is an excellent place for White's queen." reek of original research. Cite "this is good/bad"-type statements to experts.--SeizureDog 23:47, 10 February 2007 (UTC)[reply]

Proposed merge[edit]

Ok, I can accept that the series itself is reasonably notable, and so warrants coverage, however, the individual games do not need their own articles. They can easily be covered in one single page. And oddly, there doesn't even seem to be a combined page on the events, so obviously that page will have to be constructed, though I'm uncertain as to the best choice for a name. FrozenPurpleCube 21:04, 13 April 2007 (UTC)[reply]

  • There is plenty on this match in the May 1996 issue of Chess Life, and plenty more on the rematch in the "Special Summer" edition of the same publication. I would support having an article each on both matches, but there is a lot of work to do in making that, and there are many other things I would like to do. However until we get such articles, a merged item like what you did will just look like an incoherent mess of loosely stitiched together games. Sjakkalle (Check!) 14:29, 16 April 2007 (UTC)[reply]
I'm going to explain in more detail why this should be merged, by using some analogies. Let us take the World Series. There are articles on individual years like 1989 World Series or 2003 World Series not on the individual games in the series. This is also true for the regular season baseball games. Coverage of the season, not individual articles on the games. Yet all of these events receive repeated and unambiguously notable coverage.
Beyond that, there wasn't even a page on the overall Kasparov/Deep Blue matches. Is there not a problem in that itself? Specific coverage of the matches is one thing, but they did nothing to put things into a bigger picture. In short, it's just a bad way to do things and less helpful to the individual user. I fully intended to fix up the page to present a more coherent picture, I even said so in the edit summary. So your objection that the page didn't look good when merge doesn't sway me. It was nothing more than the contents as they were originally, and since I intended to fix that problem, I don't see why you're objecting. If you want a separate page on both series, I can accept that, but the current setup, with a few games in the series (note, not all of them, just a few), is not acceptable to me. If you don't want to clean it up, I will. It might not be overnight, it might take to the end of the week or longer, but I ask that you give me time to do it. If that isn't acceptable to you, I'll seek a third opinion. FrozenPurpleCube 14:36, 16 April 2007 (UTC)[reply]
I really think you should know a bit about the subject before you start trying to clean it up. Sjakkalle (Check!) 14:41, 16 April 2007 (UTC)[reply]
Basically, we cannot merge these articles until we have a suitable place to merge them to. Sjakkalle (Check!) 14:48, 16 April 2007 (UTC)[reply]
Please try to convince me with ways besides telling me I am ignorant. That is not persuasive. You simply can't assert "I know better" to get your position done. And frankly, given the state of the articles now, I don't see that there's any loss to them being in one place versus not. FrozenPurpleCube 15:00, 16 April 2007 (UTC)[reply]
And since I don't expect to convince you of anything, I'm going to seek a third opinion. FrozenPurpleCube 15:00, 16 April 2007 (UTC)[reply]
Further up you compare the situation to World Series. Have you tried comparing it to inidividual Super Bowl matches, articles such as Super Bowl XVI? Sjakkalle (Check!) 15:06, 16 April 2007 (UTC)[reply]
The Superbowl is a yearly match, occurs once each year. Unlike the world series, it is a single game. See the difference? Now the NFL does have yearly playoffs that are more than one game. Guess what? National Football League playoffs, 2006-07 is the way it is done, instead of individual articles on the games. This is the same situation with the Deep Blue Kasparov matches. An individual article on each game? Not acceptable, especially when there's not even one to provide perspective on the overall series. FrozenPurpleCube
FrozenPurpleCube, on what are you basing your merge request, besides the fact that there is no page on the overall Kasparov/Deep Blue matches? I do support a general article about the match, with links like "See Deep Blue - Kasparov, 1996, Game 1 for the main article", but I think merging the individual games into one article is not a good idea. Why don't you take your time and create a new page in your own user space at User:FrozenPurpleCube/Kasparov versus Deep Blue? Then we can discuss it. --ZeroOne (talk | @) 20:08, 16 April 2007 (UTC)[reply]
On the basis that appropriate coverage of the individual games is not of the in depth level which a subject should be before being covered in its own article, nor in line with the practices of Wikipedia as a whole. It's pretty simple really. If you wish me to quote from WP:MERGE then
  1. There are two or more pages on related subjects that have a large overlap. Wikipedia is not a dictionary; there does not need to be a separate entry for every concept in the universe. For example, "Flammable" and "Non-flammable" can both be explained in an article on Flammability.
  2. If a short article requires the background material or context from a broader article in order for readers to understand it.
are the criteria I would suggest. And I think it's clear an article on Deep Blue versus Garry Kasparov is preferable than articles on the individual games which give no background on context. Do you disagree with that concept? FrozenPurpleCube 20:45, 16 April 2007 (UTC)[reply]
I think the point 2 clearly does not hold. First, I don't think that this article counts as a short one, do you? Second, all the required background material is given in the first paragraph. The second paragraph is the greatest common divisor between the articles and would form the main content of the Deep Blue versus Garry Kasparov article (which is currently redirected into this game 1). I guessed the point 1 would be argued. These subjects overlap for their first paragraph or two only, the games themselves are individual entities and do not overlap with each other. I'll quote Wikipedia:Summary style:
Summary style is based on the premise that information about a topic should not all be contained in a single article since different readers have different needs;
  • many readers need just a quick summary of the topic's most important points (lead section),
  • others need a moderate amount of info on the topic's more important points (a set of multi-paragraph sections), and
  • some readers need a lot of detail on one or more aspects of the topic (links to full-sized separate articles).
So the Deep Blue versus Garry Kasparov article should explain the backgrounds and the current detailed articles about individual games should be left like they are, linking back to the summary article. --ZeroOne (talk | @) 21:23, 16 April 2007 (UTC)[reply]
OK, this is what I had in mind: User:ZeroOne/Deep Blue versus Garry Kasparov. Please comment on and/or improve it. What do you all think? When/If this is moved to the main namespace, the individual game articles will be edited to link back to this one and redundant background information should be removed from them. --ZeroOne (talk | @) 21:55, 16 April 2007 (UTC)[reply]
Actually, I would say that this article is actually a short one, (though it is longer than the others like Deep Blue - Kasparov, 1996, Game 5 given that the vast majority of it is coverage of individual moves in the match. That might be valid content in a book on Chess, or on the series itself, but even the articles on the World Series don't cover every inning, or the ones on the Superbowls every play. Even if you believe a move by move recounting is a good idea (and I'm not completely rejecting it), summarization would at the least increase the value accessibility to everyone. FrozenPurpleCube 22:28, 16 April 2007 (UTC)[reply]
Right now, I think the style you are suggesting for the combined page is good, but I do not agree that the individual games should have pages. Going to that much detail just exceeds the provenance of Wikipedia. FrozenPurpleCube 22:28, 16 April 2007 (UTC)[reply]
Chess is different from baseball or American football. In chess, you have to know every move to know what the current position/situation is. In baseball, you can just say "player X hit a home run". In football you can say "player Y passed for a 15 yard gain", and it makes sense. In chess, you can't say "Black moved his rook to the far side of the board". It makes no sense unless it is in context of the exact position, which requires that all moves are known. Bubba73 (talk), 15:17, 17 April 2007 (UTC)[reply]
I am not sure I agree. One could easily argue that knowing every single play in a football or baseball game is as important as knowing any move in a chess game. After all, don't you need to know how the bases got loaded and 2 outs up before understanding the importance of a home run? Besides, if all the article has to say is a recounting of the moves, I'm sorry, but there's no reason for it to be an individual article. There has to be more content than that. (And yes, that is a problem with several of the articles I'm suggesting be merged. If you feel you have to include all of the moves, fine, but don't make it an article of its own just because you think the game is important. Wikipedia doesn't list the statlines from every baseball game, or every world series. Why not? Because that is indiscriminate information. FrozenPurpleCube 15:48, 17 April 2007 (UTC)[reply]
Do you think that the moves of this game are "indiscriminate information"? They aren't a "how to". Bubba73 (talk), 00:01, 19 April 2007 (UTC)[reply]
There is no deep strategy in football or baseball the way there is in chess. The strategic goals can go through the entire game, building on each position and each move. In chess, you must know the precise position in order for a move to make sense. In football, it usually doesn't matter if they get a first down on the 41 yard line or the 42 yard line. In chess it makes a big difference whether the pawn is on e6 or e7. Tou can't sensibly recount a chess game without giving every move. Bubba73 (talk), 20:33, 18 April 2007 (UTC)[reply]
The thing is that football, baseball & such sports are "analog": they can be broken into smaller and smaller pieces. You might want to give the situation play by play or even player's locations within one meter every second but then someone could argue that you need to know the location within one inch every .1 seconds to know the position. Chess, on the other hand, is "digital": the smallest element that has any meaning is clearly defined: it's just one move. Not quarter a move, not hundredth of a move, not the move 100 times a second: it's just a move. Thus it is a lot more reasonable to list a game of chess move by move than to list a game of American football play by play, since the football version is always only an approximation whereas the chess moves describe the game totally completely. --ZeroOne (talk | @) 21:40, 18 April 2007 (UTC)[reply]
Good points, it is also like continuous versus discreet. Bubba73 (talk), 23:35, 18 April 2007 (UTC)[reply]
You can recreate a chess game completely from the moves, you can't do that with football/baseball/etc. Also, very often people DO play through the moves of a famous chess game. I don't know if many people go back and try to recreate a football or baseball game. Bubba73 (talk), 23:58, 18 April 2007 (UTC)[reply]
You know, I don't think you read what I said. This is the important part: Besides, if all the article has to say is a recounting of the moves, I'm sorry, but there's no reason for it to be an individual article. There has to be more content than that. (And yes, that is a problem with several of the articles I'm suggesting be merged. If you feel you have to include all of the moves, fine, but don't make it an article of its own just because you think the game is important. Wikipedia doesn't list the statlines from every baseball game, or every world series. Why not? Because that is indiscriminate information.
Get back to me when you're willing to acknowledge that point. I'm not terribly bothered by recounting the moves of the game if you can show the game itself important. If you can't....well, that's a problem. This is even true with these games. I think one article would work just as well, as if you can't get more individual content, a merge offers the maximum information in one place, which is more of an advantage than having a bunch of stub-level pages. FrozenPurpleCube 16:03, 19 April 2007 (UTC)[reply]
PS: Actually, there are replays of many football and baseball games, with notes and records of events in the game. It's even been used to make virtual replays of the game. And for boxing, there's even been a movie based on the concept. So no, I don't agree that the difference in granularity is meaningful. Being able to be precise doesn't justify including the information. What does is significance outside the information. FrozenPurpleCube 16:03, 19 April 2007 (UTC)[reply]
So what is wrong with someone being able to replay a chess game? With all of the moves, someone can do that. It is completely non-indiscriminant. If you listed all of the possible moves of a chess game, that would be indiscriminant. Bubba73 (talk), 17:08, 19 April 2007 (UTC)[reply]

(unindent) Let's take Deep Blue - Kasparov, 1997, Game 6 as a simple example. Kasparov made a bad move on his 7th move: 7... h6. But whether h6 is a good move, a bad move, or in between depends on the exact position. And to see why that was a bad move and how Deep Blue exploited it requires the rest of the moves of the game. Bubba73 (talk), 18:29, 19 April 2007 (UTC)[reply]

Third opinion[edit]

I came here via WP:Third opinion. I haven't done this before, but I'll give it a try. I think the one big article vs. several small articles is an editorial decision best left to the editors working a particular subject, as long as the results look reasonable and there is sufficient verifiable information in each article. My first impression of the Game 1 article was quite positive. It was clear, well laid out and about the right size. I know the game but am not a chess expert. I am not able to follow a game by reading chess notation. I got a nice feel for what happened from the article. It seemed like a model for a chess game article on Wikipedia. Same for 1997, Game 6. I'd like to see 1997 Game 2 expanded. Perhaps a few more positions such as 45 mentioned at the end.

I do agree that there is a need for a summary article. It might consist of the second paragraph of the Game 1 article, a table listing each game with its outcome and significance, links to individual articles about key games, and ideally more about what has happened since. In my opinion, the complete moves of individual games should not be in the summary article as they would put off non-expert readers.

I don't think the Baseball World Series articles are controlling precedents. The needs of each game are different. And one could argue that the Deep Blue - Kasparov matches are more notable than any World Series since they raise man vs machine issues that go far beyond the game. --agr 12:20, 18 April 2007 (UTC)[reply]

Well, unfortunately, the problem is, these articles don't even discuss that issue. Perhaps that might be worth adding. FrozenPurpleCube 13:24, 18 April 2007 (UTC)[reply]
Having a summary, with the technical details of each game pushed to individual articles, might make it easier for an editor who is not a chess expert to add well-sourced material on the broader implications. Just to be clear, I'm not saying the organization you proposed couldn't work, just that this is a judgement call and having a separate article per game is a reasonable way to go.--agr 14:35, 18 April 2007 (UTC)[reply]
Indeed, it might be, but that would depend on the level of content available. However, I don't think it's the case yet. I think it'd be better to work from one page, and expand later. In any case, the lack of a over-arching summary is the biggest problem. FrozenPurpleCube 15:20, 18 April 2007 (UTC)[reply]
I agree that a summary article is needed. Why not write one?--agr 18:25, 18 April 2007 (UTC)[reply]
Well, as I said, I believe in merging the pages together, but when I tried that I got reverted before I could get a chance to get some resources and clean it up. Since I've got no interest in covering the individual games in their own articles, I'll leave the work to the people who have expressed a desire to do it that way. As long as they're willing to cover the series as a whole, my main concerns are met. Saves me the trouble of going to the library. FrozenPurpleCube 18:53, 18 April 2007 (UTC)[reply]
Well I already began writing the summary article, as I stated above: User:ZeroOne/Deep Blue versus Garry Kasparov. I'm planning to import it to the main namespace after finishing the initial copying of the moves and writing two-sentence summaries. I already suggested you should start such a fork in your own user space. Now that I've done so, we could compare the results. --ZeroOne (talk | @) 21:22, 18 April 2007 (UTC)[reply]
Good, when the summary article gets to the namespace that should settle the issue. Bubba73 (talk), 21:29, 18 April 2007 (UTC)[reply]
I'm sorry, but I have zero interest in making a page just to compare results with yours. If I don't feel the page version you have is adequate, I'll just make appropriate edits to it or make suggestions as to improvement. FrozenPurpleCube 15:48, 19 April 2007 (UTC)[reply]
I was speaking of ZeroOne's page. Bubba73 (talk), 16:32, 19 April 2007 (UTC)[reply]

Summary page[edit]

I'm finished, I consider User:ZeroOne/Deep Blue versus Garry Kasparov ready to be published in the main namespace. However, I cannot move it as there is a redirect in place of the target article, so I've just placed a move request at Wikipedia:Requested moves. In case you disagree with me, just go ahead and delete the request, fix the problem and add the request back. Or delete the request and leave a constructive comment here. If/When the article is moved, Deep Blue - Kasparov, 1996, Game 5 and Deep Blue - Kasparov, 1997, Game 2 should be changed into redirects. About referencing: if no reference number is given, all the claims are referenced to [2] (which is listed in the references) and its sub-pages. --ZeroOne (talk | @) 00:54, 20 April 2007 (UTC)[reply]

A redirect page with that name exists, but I think you can just take your text and copy it over to the new page. Bubba73 (talk), 01:19, 20 April 2007 (UTC)[reply]
I moved the text. Bubba73 (talk), 01:54, 20 April 2007 (UTC)[reply]

Is Kasparov white or black?[edit]

The text in this article is ambiguous in terms of which side is Kasparov and which is Deep Blue. Above the board it shows Kasparov as Black, however the game annotation indicates that Kasparov is white. Which is correct?

Trojanfoe (talk) 14:05, 12 September 2008 (UTC)[reply]

Kasparov is Black in this game. The comment after White's 23rd move may be confusing: "This type of pawn sacrifice is typical of Kasparov's style of play. Indeed, Kasparov commented that he might have played 23. d5 himself". It is talking about White's move, but it is a move by Deep Blue that Kasparov might have played if he were White. Bubba73 (talk), 16:42, 12 September 2008 (UTC)[reply]


Bad analysis[edit]

THe analysis here is pretty much nonsense. The game is even until 27...d4, which loses. — Preceding unsigned comment added by 71.230.169.158 (talk) 00:17, 11 January 2012 (UTC)[reply]

Raymond Keene says White probably stands better after 21...f5, and after 25.b3 says Black is still not yet lost if he plays 25...Rd8! – so, he disagrees with you. Also, do you have any GM (or even IM or Master) analysis which supports your contention that 27...d4 was the losing move? Also, the analysis in the article seems pretty reasonable to me, can you be more specific regarding what you think is "nonsense"? Ihardlythinkso (talk) 05:40, 11 January 2012 (UTC)[reply]
Well, Houdini 1.5 (34/57), which is far beyond GM strength, gives an evaluation of 0.0 after 27...f4, and White +.70 after 27...d4. For another example, the analysis gives a ? to 25...Kh8, which Houdini judges as the best move (0.0), while recommending 25...Ne7 (-.31). For another, 25. b3 is given an exclamation, but Houdini judges it as .2 inferior to 25. b4. So most of the punctuated moves are misspunctuated, the blunder that lost the game is unpunctuated. I agree that most of the comments themselves are pretty reasonable (except for the one on move 25), but because text never says anything like "this is the fatal error", the effect of the mispunctuation is to make a reader believe that black lost because of the ? moves, which is not the case.
All of which raises a key question: can analysis cited to a published piece of software (such as Houdini) with excellent bona fides be considered verifiable under wikipedia guidelines? Erniecohen (talk) 17:03, 11 January 2012 (UTC) Erniecohen (talk)[reply]
That is a very good question, and it needs to be addressed beyond just this article. It seems like a grey area to me. My feeling is that it would be regarded as original research if an editor did it, but maybe not. It also might be a reliable source and verify issue, since there are so many variables - the version of the software, what computer it runs on, how much time it is allowed to run, etc. Bubba73 You talkin' to me? 18:19, 11 January 2012 (UTC)[reply]
I agree with Bubba, but see the distinction more clearly (i.e., not a grey area): if Houdini's analysis is published in a reliable source, it's fine to include, otherwise no (it's original research). Sasata (talk) 18:27, 11 January 2012 (UTC)[reply]
Right, I don't see any problem if it is published in a reliable source - only if the editor does it. Bubba73 You talkin' to me? 18:42, 11 January 2012 (UTC)[reply]
I agree that the analysis is pretty much nonsense (not precise or accurate) but it is what we have from sources. It's the same all over Wikipedia. Regards, SunCreator (talk) 00:21, 12 January 2012 (UTC)[reply]

I also agree that the analysis is bad. — Preceding unsigned comment added by Thebestwikipedian (talkcontribs) 21:47, 14 September 2015 (UTC)[reply]

Quote about game[edit]

Charles Krauthammer wrote a comment on move 29. Nxb7: "Deep Blue's king was under savage assault by a World Champion. Any human Any human under such assault by a world champion would be staring at his own king trying to figure out how to get away. Instead, Deep Blue ignored the threat and quite nonchalantly went hunting for lowly pawns at the other end of the board. In fact, at the point of maximum peril, Deep Blue expended two moves - many have died giving Kasparov even one - to snap one pawn. It was as if, at Gettysburg, General Meade had sent his soldiers out for a spot of apple-picking moments before Beckett's charge, because he had calculated that they could get back to their positions with a half-second to spare. In humans, that is called sangfroid. And if you don't have any sang, you can be very froid. But then again if Meade had known absolutely - by calculating the precise trajectories of all the bullets and all the bayonets and all the cannons in Pickett's division - the time of arrival of the enemy he could, indeed, without fear, have ordered his men to pick apples. Which is exactly what Deep Blue did. It had calculated every possible combination of Kasparov's available moves and determined with absolute certainty that it could return from its pawn-picking expedition and destroy Kasparov exactly one move before Kasparov could destroy it. Which it did. It takes more than nerves of steel to do that. It takes a silicon brain. No human can achieve absolute certainty because no human can be sure to have seen everything. Deep Blue can."

This appears in https://en.chessbase.com/post/25-years-ago-deep-blue-beats-kasparov and partially in https://www.journalstandard.com/story/opinion/columns/guest/2011/03/15/my-view-education-must-appreciate/44912266007/ (And was memorable enough that I thought of it 25+ years later.) Does anyone have the original citation?

Jrincayc (talk) 15:52, 5 February 2023 (UTC)[reply]

Ah, found it at https://content.time.com/time/subscriber/article/0,33009,984175-2,00.html and added it to wikiquote. Jrincayc (talk) 16:43, 5 February 2023 (UTC)[reply]